Inductive link for a medical implant
The present invention relates to an inductive link for power and signal transmission to a medical implant, pref- erably power and signal transmission through the skin or any other physical barrier of a human being or living animal having a medical device implanted in the body.
Inductive signal and power transmission through the skin is well known in the art and the technique has been used for instance for nerve and muscle stimulation, for im- plantable hearing aids and other implants. Inductive transmission is preferred instead of other alternatives such as implanting a battery or breaching the skin by a physical canal. Implanting a battery provides a potential risk as a battery often involves substances which might be injurious to the body and a physical canal through the kin could be a serious disadvantage due to potential infection in and around the canal.
For stimulating nerves and muscles some type of electronic equipment is normally provided in the body tissue. Such an electronic equipment requires power as well as some kind of information for its function. The most conventional way to provide the implanted electronic equipment with the necessary power and information has been through inductive signal transmission. The principle for such a signal transmission is two coils, one of the coils placed on the inside of the skin and the other coil on the outside. An alternating current is fed through the the outer coil, the so-called primary coil. The alternating current then generates an alternating field where a part of the field goes through the inner, secondary coil. In the secondary coil a voltage is induced which is rectified so that a direct current is generated on the inside of the skin. The information can be transmitted by varying, or modulating the primary field in some way. Either the amplitude is modulated, i e amplitude modulation, or the frequency is modu-
lated, i e frequency modulation. And by de-modulating the carrier signal on the secondary side the power content and the information content in the signal can be separated.
The magnitude of the voltage which is induced on the inside of the body depends among other things on the voltage on the outside, the coil turn ratio, how much of the primary field that goes through the secondary coil, possible voltage drops ec How much of the Ω imar'" fiel that co s through the secondary coil depends on the size of the. coils the distance between, the. cαils the geometrical design of the coils and possible variations in the permeability between the coils. The coupling coefficient is defined as the ratio between the field generated by the pri- mary coil and the field enclosed, by the secondary coil.. The coupling coefficient lies between zero and one. The higher coupling coefficient, the higher efficency as the primary current then is reduced for a certain secondary voltage which gives less resistive losses on the primary side. To achieve a high coupling coefficient the distance between the coils has to be sufficiently small compared to the diameters of the coils. But for a certain distance and a certain maximum coil diameter there are other ways to increase the coupling coefficien . The most commonly used approach is perhaps to dispose ferrite material in the center of the coils and thereby forming the field so that more of the primary field, is σirected through the secondary coil. Unfortunately, in a ferrite material there are also eddy currents induced which currents generate losses so it is not sure that the increased coupling coefficient really means an increased efficiency. In US 5,891,183 it is described a method to increase the coupling coefficient by using coils of spiral shape. The reason for an increased coupling coefficient in this case is the fact that the spiral shape of the coils concentrates the field more to the center of the coil.
Voltage drops between the power source on the primary side
and the load on the secondary side consist of resistances in driver circuits, resistive components in the, coils and distributed inductances in the coils . A voltage drop in the driver circuit can be reduced by using driver transis- tors with a low forward voltage drop. However, this means a higher driving current and slow switching, times so that the choice of driving transistors must be balanced against this fact. The distributed inductances in the coils can be extinguished by coupling a capacitor in series or in par- allel w th the coil in order to make the coil resonant at the operating frequency. The frequency of the carrier wave can be varied in each specific case. It can be said that in order not to increase the size of the coils too much the operating frequency should be some hundred kHz or more. However, if the frequency is too high the coils become gradually self resonant and then more σapacitive than inductive. This happens for frequencies in the range of 10-20 MHz depending on how the coils are wound.
An inductive power and signal transmission link of the type which has been described so far is usually composed of an external power source, a driver circuit and a primary and secondary coil. The external power source is typically a battery. Unfortunately, the operating time of a battery and the ability to deliver power is limited and the battery has to be exchanged at regular intervals. It- is of course desirable to reach a high overall efficiency between the battery and the load as this means an increased operating time and a higher amount of power on the load. The electronics beneath the skin often requires a relatively even voltage supply. As previously mentioned however, the amplitude of the induced voltage on the secondary side varies if the coupling coefficient is changed which happens if there is a relative displacement between the primary and secondary coils. Even a change of the load on the inside of the body affects the voltage due to the voltage drop over the link and some other series resistances between the battery and the load. In order to ob-
tain a supply voltage on the secondary side which is relatively independent of variations in load and positions some type of voltage control is required. Such a voltage control can be achieved in different ways. For instance a feedback signal could be sent back in some way from the secondary side to the primary side where the primary field is adjusted in such a way that the secondary voltage is kept within a desired interval. Such a type of voltage control is described in US 5,876,425 and US 6,442,434.
In US 5,735,887 another type of control is described in which a loading pulse signal is generated when the voltage on the secondary side falls outside a certain interval. This loading pulse signal is sensed on the primary side where also the control is performed.
Another type of voltage control is described in US 5,562,714 and US 5,117,825. In this case the voltage control is accomplished by means of a sensor coil which is arranged on the primary side and which senses the primary field for controlling it towards a predetermined value. By this method a control of variations in the load is obtained but there is no control of variations in the position.
In US 4,679,560 it is described a method to achieve a stabilised voltage by means of stagger tuning the primary and secondary side on each side of the operating frequency. The stagger tuning has the effect that if there is a reduction of the relative distance between the coils the poles are moved away from each other due to the inductive coupling. This pole separation reduces the voltage which counteracts a voltage increase which otherwise should have happened for such a reduction of the relative distance be- tween the coils. This type of stagger tuning is also described in US 5,070,535. A common feature for both of these patents is the requirement of a separate oscillator for generating the operating frequency.
In an inductive link some type of retaining element is required which guarantees that the outer and inner units are hold together. The most commonly used method is to use re- taining magnets. And in order to limit the outer dimensions of the outer and inner units as much as possible the magnets are typically arranged in the center of the coils. However, eddy currents are induced in the magnets which give rise to resistive losses. In US 6*17.8*353 it Is dse- cribed a method to laminate the retaining magnets in order to reduce the eddy current losses.
It is an object of the present invention to provide an inductive link for power and signal transmission to a medi- cal .implant which is substantially simpler and with less power demand compared to prior links of this type. The inductive link should provide an efficient transmission of the power as well as the signal at a comparatively low voltage and power supply level, i e within a power range of a few mW. It should also contain less active electronics in order to limit the current demand.. A further demand on the inductive link is the ability to provide voltage, that is independent of variations in the load and variations in the relative positions in a longitudinal as well as in a side direction. According to the invention two coils are utilized which are orientated in space in such a way that they are located on each side of a physical barrier, specifically the skin of the human bod - One of the coils, the so-called primary coil, is located outside the physical barrier and is a part of the primary circuit that generates an electromagnetic field. The other coil, the so-called ssecondary coil, is located on the other side of the physical barrier and is a part of the secondary circuit which can absorb and utilize the entire or a part of the magnetic field emitted by the primary coil. The primary and the secondary coils has a mutual coupling coefficient k which is less than one. The coupling coefficient depends on the relative position between the coils which
can be varied by time.
A new and characterizing feature of the invention is the introduction of a sensor circuit in the form of a struc- ture of electrical wire conductors arranged in a physical close relationship to the primary coil in space so that a part of the magnetic field from the secondary coil is enclosed by the sensor circuit thereby providing an induction in the sensor circuit which can be detected and thereby indicate the condition in the secondary circuit.
A further characterizing feature of the invention is that the sensor circuit is orientated in space in such a way that the part of the magnetic field from the primary coil which is enclosed by the sensor circuit does not give rise to any induction in the sensor circuit, which means that the sensor circuit only senses the condition in the secondary circuit. This is obtained when the integral of the magnetic field generated by the primary coil and enclosed by the contour of sensor circuit goes towards zero.
According to a preferred embodiment of the invention the sensor circuit is arranged to sense the instantaneous voltage on the secondary side, i e both the phase and the amplitude of the voltage.
According to a further preferred embodiment of the invention the sensor circuit is fixed mounted on the primary coil without any inductive coupling between the coils and the sensor circuit further comprises one or more individual coils.
In contrast to prior signal and transmission links the feedback circuit is in this case completely passive. There is no electronics on the secondary side that returns any information to the primary side, instead it is the primary side that senses the voltage on the secondary side by means of said sensor circuit. By the feed back of the
phase of the secondary tank circuit and the control of the primary side with a certain phase value relative to the phase of the secondary side a self oscillating signal transmission is obtained. By the feed-back of the voltage of the secondary side any variations in the voltage caused by a change of load or position can be compensated for by adjusting the voltage on the primary side.
By means of said feed-back arrangement a closed loop is obtained which makes it possible to keep the voltage constant on the secondary side by controlling the input power on the primary side. As the sensor coil only senses the field and thereby the voltage on the secondary side and this is accomplished by means of completely passive elec- tronics there is no additional current consumption for the feed-back which means a very efficient way to control the voltage.
In the following the invention will be described more in detail with reference to the accompanying drawings which illustrate some different embodiments of the inductive link according to the invention, wherein
Figure 1 illustrates by means of a block diagram the sep- arate parts of an inductive link for signal and power transmission,
Figure 2 illustrates by means of an electric circuit diagram the separate components more in detail,
Figure 3 illustrates schematically the position of the coils in space,
Figure 4a shows the relation between Rp, Rα and R when cal- culating the field around a coil,
Figure 4b shows the field characteristics as a function of the radius of coils which are wound in two different ways,
Figure 5 illustrates a primary coil with a sensor coil arranged in such a way that the total field from the primary coil through the sensor coil is zero,
Figure 6 shows the sensor characteristics in the longitudinal direction between two different pair of spoils,
Figure 7 illustrates some examples of the physical config- uration (wire structure) of the sensor coil,
Figure 8 illustrates a pancake coil wound by itz wires,
Figure 9 illustrates a sensor coil which is insensitive to lateral displacements,
Figure 10a shows sensing charactersistics in the radial direction for three different coil distances d,
Figure 10b illustrates two coils on a distance d from each other, and
Figure 11 shows an example of the physical configuration of an inductive transmission link according to the inven- tion.
Figure 1 illustrates by means of a block diagram the individual components of an inductive link for power and signal transmission. The system comprises a primary side 1 and a secondary side 2. The primary side 1 and the secondary side 2 are physically separated by a physical barrier 3, for instance the skin of a human being. The primary side comprises a transmitter 4, a driver circuit 5 and a power voltage supply 6. The purpose of the transmitter is to transform an alternating current to an alternating magnetic field while the driver circuit 5 by means of a control signal controls the current through the transmitter as desired. On the secondary side the magnetic field is
again transformed into electrical energy in a receiver 7. In order to drive a load 12 the voltage signal is converted into a rectified voltage in a converter 8. In order to maintain a constant voltage on the secondary side 2 there is a feed-back of the voltage on the tank circuit on the secondary side of the system to the primary side 1 on which side the input power can be controlled. The feedback is accomplished by means of a sensor circuit 11 which senses the field from the tank circuit on the secondary side and thereby the instantaneous voltage. It is possible to send information on the carrier wave by modulation which is carried out in the circuit 9. The most simple way to achieve a modulation of the carrier wave is to vary the power supply voltage on the primary side. If this is the case then the carrier wave has to be demodulated on the secondary side which is indicated by the circuit 10. The rectified voltage from the rectifier 8 and the demodulated signal from the circuit 10 provides the load 12 with power as well as signal information. Generally the modulation of the carrier wave can be performed in three ways, i.e. by means of amplitude modulation, by frequency modulation or by phase modulation. As it is important to keep the power consumption as low as possible amplitude modulation is used in this case.
In figure 2 the system is described more in detail by means of an electric circuit diagram. The generation of the magnetic field on the primary side 1 is effectuated by means of a coil 15, the so-called primary coil. Current supply to the coil 15 is provided by two transistors 16, 17 and a comparator 18. On the secondary side 2 the magnetic field is transformed into electrical energy by means of a secondary coil 19. The coil 19 and a capacitor 20 provides a parallel LC circuit which is tuned to the car- rier frequency in order to improve the transmission capacity. In order to drive the load the voltage is rectified by means of a diode 21. The current is smoothed by means of a capacitor 22 so that a constant voltage is obtained.
A diode 23? a capacitor 24 and a resistor 25 are forming an envelope detector which is a basic demodulator for an amplitude modulated system. The rectified power voltage and the demodulated signal provides the loa 26 wit eπer- gy as well as signal in ormation. As mentioned above the amplitude modulation is carried out by varying the power supply voltage to the driver transistors 16 and 17, This can be accomplished by a transistor 27 « In order to maintain a constant voltage on the secondary side there is a feed-back of the voltage of the tank circuit an the secondary side to the primary side where the input power can be controlled lay means of an operating amplifier 28 an a transistor 27. The feed-back is accomplished by a coil 29 which senses the field from the tank circuit on the seeon- dary side and thereby the instantaneous voltage.
The coil 29 is a completely passive sensor coil by which the field on the secondary coil 19 is fed back to the primary side. By this feed-back arrangement a closed loop circuit is obtained which makes it possible to maintain a constant voltage on the secondar side by controlling the input power on the primary side. Furthermore, the feedback phase is use to control the primary driving which eliminates the need of an external oscillator. In order to obtain a correct phase position for the driving on the primary side there is a phase-shifting filter included in the circuit between the feed-back voltage and the primary driving. The filter comprises in this case a resistor 30 and capacitor 31. in order to exclude any contribution on the senso coil 29 from the primar coil 15 the senso coil is located above the region in which the field from the primary coil is changing its sign. If the senso coil is located in such a way that the positive field and the negative field which are enclosed by the sensor coil are equal, then the contribution from the primary field is zero., which will be described more in detail in the following. The secondary field on the other side will be indicated in the sensor coil. As the sensor coil 29 only sens-
es the field and thereby the voltage on the secondary side, and this is accomplished by entirely passive electronics, there is no additional current consumption for the feed-back which makes this arrangement a very effi- σient way to control the voltage and the primary driving.
In figure 3 it is schematically illustrated how the different coils are orientated in space. The primary and the secondary coils, 15 and 15 respectively,- are arranged on a. certain distance from each other which distance is changed in the course of time- The sensed voltage in the secondary coil 19 will of course depend on the distance between the secondary coil and the sensor coil 29- However, by using a suitable geometry on the primary and the secondary coils it is possible to find a maximum of the sensed voltage for a certain coil separation. And if this distance is arranged to the center of the distance interval between the primary and the secondary coil then a sensing characteristic can be obtained which is relatively insensitive to σhanges in the coil separation.
The sensor coil is located in a close physical relationship with the primary coil preferably the sensor coil is fixed mounted on the primary coil without any magnetically coupling between the coils. In the most simple case the sensor coil has a radius which is substantially less than the radius of the primary coil. The sensor coil has a peripherical location relative to the primary coil above the region where the field from the primary coil changes its sign, see more below.
The coils are hold together in a known way, for instance by means of retaining magnets* The magnets are preferably located in the peripherical region of the coils as the low field concentration in these regions only generates very small eddy current losses. The location of the magnets in the peripheriσal region of the coils also gives a more preferred field configuration between the inner and outer
units of the link, which reduces the risk for the outer unit to loosen.
The principle for the feed-back is something that makes this invention quite different from previously known signal and power transmission links. Prior feed-back arrangements have all been based on some type of active coupling, i e by means of active electronics provided on the secondary side and which generates some type of status signal which is sensed by the primary side. In the present σaase, however, there is no electronics whatsoever on the secondary side for providing any status signal. On the other hand, the problem with a sensor coil on the primary side for sensing the field from the secondary side is the unde- sired contribution from the primary coil 15 and the sensitivity to changes in coil separation. The undesired contribution from the primary coil can be avoided by placing the sensor coil in such a way that it encloses a field from the primary coil which always has two signs and that the sum of these two fields is zero. In order to illustrate how this is possible the field from a coil will be described in the following:
The field strength in a point on the distance R from a wire conductor having a length of dl is defined as: dg ^ μ0i dϊxR 4π R*
where μo = permeability constant in vacuum, i = current strength through the wire, —> R = distance vector between the wire and the point in question, and dl = direction vector for a wire element dl.
The integral of dB for the entire length of the wire gives the field strength for a specific point.
For a circular loop we have:
An φ{Q Rp 2 + RC 2 - 2RcRp cos φ Rp 2 + RC 2 - 2RcRp cos φ
" where Re, Rp and cos ? are illustrated in figure 4a.
The most easy way to solve the above integral numerically is to use for instance MAT AB.
Figure 4b shows the field characteristics as a function of the radius of coils. In the figure it is also illustrated characteristics for coils which are wound in two different ways, i e an outer radially wound coil (this means that the wire turns are concentrated to the peripherical region of the coil) and a spiral shaped coil (this means that the wire turns are spread along the radius of the coil, a so- called pancake coil). The field strength is normalised to the field strength in the center of each coil and the ra¬ dius is normalised to the outer radius of the coil. The reason that the theoretical characteristics does not fully correspond to the measured characteristics depends on the fact that the diameter of the wire conductor in this σal- culation has been approximated to zero and also on the fact that the testing coil which has been used for sensing the field has a certain diameter which makes it impossible to sense the field in the immediate vicinity of the wire. It is shown in figure 4b that for both of the two coil types the field is changing its sign approximately at the outer radius.
In figure 5 it is illustrated a primary coil 15 with a
sensor coil 29 located in such a way relative to the primary coil that the total field from the primary coil through the sensor coil is zero. In the figure the primary coil 15 is wound as a pancake coil with all the wire turns in one plane. However, this is not necessary to achieve the zero effect. The important thing is that the field representation always has two different signs at the same time. However, the location of a sensor coil on a primary coil of a type having the wire turns arranged in the outer peripherical region of the coil requires a much higher precision. This depends on the fact that there is a much more abrupt zero passage for these types of coils, see figure 4b. How the generated voltage in the sensor coil 29 depends on the separation between the sensor coil 29 and the secondary coil 19 in turn depends on the geometry of the two coils. The best result with respect to sensitivity to changes in coil separation is obtained if there is a maximum for the sensed voltage somewhere in the middle of the actual distance interval between the primary side 1 and the secondary side 2. This can be obtained by a suitable geometry.
In figure 7 it is illustrated some examples of the physical configuration (wire structure) for the sensor circuit 11. In the examples above the sensor circuit has been described in the form of a sensor coil 29 which has an asymmetrical loction above the primary coil as shown in figure 7a. The disadvantage by having only one sensor coil as shown in figure 7a is the fact that there is an obvious sensitivity to lateral changes between the coils for this case. By using two coils as illustrated in figure 7b the sensitivity for changes in the y-direction is substantially reduced. The voltage signal obtained in this case is the sum of two partial voltages u-_ and u2. With two additional coils as illustrated in figure 7c the same improvement is also obtained in the x-direction. Even for this case the obtained voltage signal is the sum of all the
partial voltages, i e the sum of uχ, u2, u3 and u4. So the higher number of sensor coils, the less variation of the sensitivity in the lateral direction is obtained. A further advantage by using more sensor coils is the fact that the signal level is increased which means an improved signal to noise ratio. If all these small coils are coupled together into one single coil, then a so-called horseshoe sensor coil is obtained, as illustrated in figure 7d. In this case it is obtained a sensor characteristic that is quite insensitive to any changes in the angle, apart from the small gap formed between the two outer edges of the horseshoe. However, a substantial disadvantage with the horseshoe coil is the fact that it is rather difficult to manufacture. Finally, in figure 7e it is illustrated a further example of the sensor circuit which might be the most preferred embodiment. In this case the sensor circuit comprises two coils having different coil radius and cen= trally located around the central portion of the primary coil where the sensed voltage is the difference voltage of u1 and u2.
Preferably, at least one of the primary and secondary coils is formed as a flat, pancake coil. Also the sensor coil could have such a shape. A pancake coil provides a slimmed physical construction of the transmission link but also the most favourable coupling coefficient, because it has turned out that a pair of pancake coils, i e coils with radially wire turns, provides a substantially higher coupling coefficient compared to a pair of coils with more spread turns, i e axially wound turns. In figure 8 it is illustrated a pancake coil with Litz wires, i e multi-unit wires which consist of a plurality of tiny, individual conductors, and in which at least one of these conductors has a thermal glue coating. It has turned out that pancake coils wound by Litz wires impregnated with thermal glue provide sufficiently high Q-values (quality factors). This method also makes it possible to manufacture the coils in
a more simple way.
By having the coils wound in spiral shape and using a multi-unit wire composed of a plurality of small conductors coated with a thermal glue composition as described above a number of advantages is obtained compared to the use of a homogenius wire conductor or having all the coil turns wound at the outer diameter of the coil. The spiral shape makes the field more concentrated to the central part of the coil while it is decreasing towards the outer peripherical region. The concentration of the field provides an increased coupling coefficient and the reduction of the field towards the outer peripherical region makes it possible to place such details here in this region which are unsuitable for a location in a strong magnetic field, for instance the retaining magnets that has been described A o .
Preferably the sensor coil has a winding according to fig- ure 7e. This kind of winding with multi-unit Litz type conductors is illustrated more in detail in figure 9. Figure 9 also illustrates how the coils are connected to each other in order to obtain the zero field effect from the primary coil. This geometrical design is suitable physi- cally as well as electro agnetically for use with the pancake coil in figure 8. Provided that the coils are substantially plane parallel a design according to figure 9 result in a sensed voltage which only depends on displacements in the axial and radial direction which means that it is insensitive to changes of rotation.
Figure 10a shows measured sensing characteristics in the radial direction for three different coil distances d as illustrated in figure 10 b» As illustrated in figure 10a a radial variation of ±2.5 mm and an axial variation of ±3 mm gives a variation of the sensed voltage of only ±5 %. The outer radius of the sensed coils are in this case approximately 12 mm.
Figure 11 shows an example of the physical configuration of the inductive transmission link. The transmission link comprises an external part 1" and an internal part 2' which parts are separated by a physical barrier 3, for instance the skin of the body. In the external part there is a coil member 30, a magnet device 31, electric circuits 32, a battery 33 and some input signal 34. In the internal part 2" there is a coil member 35, a magnet device 36, electric circuits 37 and some output signal 38 (stimuli).
The invention is not limited to the examples which have been illustrated here, but can be varied within the scope of the accompanying claims. For instance, the coils might be manufactured according to any litographical method of the type which is previously known from the printed circuit card technology or semiconductor industry, preferably in the form of a multilayer method.